DESCRIPTION: This project will produce estimates of active life expectancy (ALE; the years of life expected to be lived without serious disability) using multidimensional stochastic process population models applied to, initially, the 1982, 1984, 1989 and 1994 National Long Term Care Surveys (NLTCS) and mortality data for the period 1982 to 1997. The NLTCS is nationally representative of the U.S. elderly population both cross-sectionally, for each NLTCS measurement, and longitudinally. In addition, ancillary processes of changes in ALE will be estimated from Medicare health service use data. To produce ALE estimates it is necessary to develop new multidimensional stochastic process models appropriate to analyzing partly observed processes to accurately describe changes in the health and functioning for the total U.S. elderly population and its important demographic components, e.g. gender, race, cohort, and education group. When the 1999 NLTCS is completed in the second year of this project (the NLTCS is being collected in another, already ongoing, data collection project) the analyses will be extended to a fifth NLTCS wave and to the year 2001 for mortality and service use. The investigators believe that this will greatly enhance their ability to study changes in the health of a) the extreme elderly population, b) specific birth cohorts--both before and after recent (1997) changes in Medicare programs have been instituted. The 1999 NLTCS will also help the investigators examine the prediction of ALE changes based on the 1982 to 1994 NLTCS and mortality data to 1997. Studies will also be made of the influence on functional changes of co-morbid processes. Analyses of ALE will not be restricted simply to the decomposition of life expectancy after a given age into active and inactive states. Using the detailed measures of disability made in each NLTCS wave, the project will examine more refined changes in specific types and intensity of disability and differences over age, race, gender, and education.